Hot Tweets: Stock Advice in 140 Characters or Less
CREDIT: Twitter image via Shutterstock
People looking for a new way to invest may want to look no further than their Twitter feed. That's because new research has found that investments based on data from Twitter outperformed other investment models and the Dow Jones industrial average as a whole during a four-month sample.
To arrive at these findings, the researchers looked at the number of tweets about a given company and the way those tweets are linked to other relevant topics and users. The research found that the number of tweets related to things such as the hiring of a new CEO or a new product launch within a company had a slight correlation to the stock price and the number of trades.
Lead author Vagelis Hristidis, an associate professor at the University of California-Riverside's Bourns College of Engineering, and co-authors Eduardo Ruiz, a graduate student at the university, along with Carlos Castillo, Aristides Gionis and Alejandro Jaimes, employees at Yahoo! Research in Barcelona, tested this by looking at 150 randomly selected companies in the S&P 500 Index for the first half of 2010. The researchers then simulated investments and analyzed their findings.
During the time of their investments the Dow Jones industrial average fell by 4.2 percent overall, but investments based on the Twitter data fell only 2.4 percent. The losses in other models ranged from 13.1 percent to 3.8 percent.
"These findings have the potential to have a big impact on market investors," Hristidis said. "With so much data available from social media, many investors are looking to sort it out and profit from it."
Hristidis, however, still wishes to test the model in a market where the Dow Jones increases to see if the model works in both up and down markets.
This research was presented last month at the Fifth ACM International Conference on Web Search & Data Mining in Seattle.
Reach BusinessNewsDaily staff writer David Mielach at Dmielach@techmedianetwork.com. Follow him on Twitter @D_M89.